DAT 370 - Designing and Implementing a Data Science Solution on Azure
Course Description
Learn how to train, manage, and deploy machine learning models on Azure. The course focuses on Azure Machine Learning, exploring the service and features like assessing data, managing compute, tracking the training machine learning models, implementing Responsible AI principles, and deploying models to endpoints.
Work with notebook and scripts to train machine learning models and use Azure Machine Learning managed compute for workloads.
Students are expected to be familiar with the basic data science and machine learning concepts.
This course covers the objectives for Microsoft Exam DP-100: Designing and Implementing a Data Science Solution on Azure.
University of Calgary is Microsoft Education Global Training Partner.
Course Details
Learning Outcomes
By completion of this course, successful students will be able to:
- Understand the Azure Machine Learning workspace resources and assets.
- Provision an Azure Machine Learning workspace through the Azure portal or Azure CLI.
- Create and manage data assets and datastores within the workspace using the Python SDK.
- Create and manage compute resources within the workspace using the Python SDK.
- Train a machine learning model with the no-code Designer in the Azure Machine Learning Studio.
- Use Automated Machine Learning to explore featurization and algorithms.
- Train and track a machine learning model in a notebook in the Azure Machine Learning workspace.
- Train and track a machine learning model using scripts in the Azure Machine Learning workspace.
- Create and schedule Azure Machine Learning pipelines.
- Deploy a machine learning model to a real-time endpoint.
- Deploy a machine learning model to a batch endpoint.
- Apply Responsible AI principles to data, models, and model training.
- Monitor data and models.
Topics
- Getting Started with Azure Machine Learning
- No-Code Machine Learning
- Running Experiments and Training Models
- Working with Data
- Working with Compute
- Orchestrating Machine Learning Workflows
- Deploying and Consuming Models
- Training Optimal Models
- Responsible Machine Learning
- Monitoring Models
Notes
This course includes hands-on activities to reinforce the concepts taught and provide a practical learning experience.
Lab access will be provided at no additional cost.
Prerequisites
No mandatory prerequisite.
Self-assessment for enrolment:
A minimum of 6 months relevant working experience and knowledge in:
- Basic data science and machine learning concepts
- Training and deployment of machine learning models using Notebook
OR
Recommended prerequisites:
- ICT 779 Python for Data Analysis
- DAT 120 Practical Machine Learning for Business
Applies Towards the Following Program(s)
- Machine Learning and AI on Microsoft Azure : Required